Project Summary Interventions designed to delay or prevent Alzheimer's disease (AD) are more likely to be effective if delivered early in the course of the disease before clinical symptoms emerge. To identify individuals who are in the pre- symptomatic or preclinical phase of AD the processes underlying early pathogenesis must be better understood. Recently developed tau-specific positron emission tomography (PET) tracers are allowing researchers to shed light on this critical neuropathology, which previously could only be examined post-mortem or by proxy measures from a spinal tap. Braak staging, a framework for staging tau pathology in post-mortem tissue based on density and topology of tau, may now be used to stage tau pathology in vivo. The current proposal is designed to elucidate the temporal dynamics of AD pathology, especially tau, and associated functional connectivity changes in cognitively healthy older adults. Specifically, longitudinal PET imaging of both beta-amyloid and tau along with task-free functional MRI will be used to 1) characterize the extent and spread of tau pathology using in vivo Braak staging, 2) examine the effect of tau pathology in Braak regions on local functional coherence and 3) determine how tau pathology in medial temporal lobe (MTL) affects functional connectivity to the retrosplenial cortex, which is anatomically connected (mono-synaptic) to MTL and essential to normal memory function. The relationship of tau pathology, functional connectivity changes and early changes in memory function will also be explored. The progression and severity of tau pathology are associated with disease severity (cognition) across the mild cognitive impairment (MCI)-dementia spectrum. It is critical, therefore, to elucidate the temporal dynamics of tau progression in aging and to examine possible mechanisms underlying the link between tau and cognitive decline. In this proposal both local and distant functional connectivity will be explored to better understand the downstream effects of tau on brain function. The Braak staging framework, which was originally derived from cross-sectional data, is an essential component of neuropathological diagnosis of AD. The current proposal will apply a recently developed in vivo Braak staging approach to longitudinal data for the first time. Findings from the studies described here will be novel and high impact. Furthermore, the outcomes of this proposal will contribute to efforts to differentiate normal aging from pathological aging-related trajectories. The accurate identification of individuals with preclinical AD will be critical to effective delivery and assessment of early intervention approaches. 1